=== dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [01:53] Hm. My battery indicator applet remains at 84% forever [01:54] ( well, until it actually dies really ) [01:57] genii: Is this a change from previous behaviour? I have a laptop that does something similar (79%) because of a hardware bug. [01:58] persia: It worked perfectly until I did the daily dist-upgrade today [01:58] Oh, good. A fixable bug. [01:59] persia: You have bug number to add to for the 79% one? [01:59] No. I didn't file a bug, because the HW is broken. Not a software issue. [02:00] Ah, OK [02:00] persia: Acer? Just curious [02:01] Sharp PC-Z1: first production run. It's fixed in the new model in the shops, but I'm cheap. [02:01] OK, thanks === JontheEchidna changed the topic of #kubuntu-devel to: Kubuntu <3 Channel | Test ISO images! | 10.04 LTS RC released!, Finally frozen | https://wiki.kubuntu.org/Kubuntu/Todo | milestoned bugs marked as kubuntu http://tinyurl.com/yjybcx9 Only 3 left to fix! === dendrobates is now known as dendro-afk === keffie_jayx_ is now known as keffie_jayx [06:25] apachelogger: ping [06:26] apachelogger: hey this is shadeslayer [08:23] ScottK: when you wake up, I just wanted to check how the backports are going? [09:37] The default audio player in the KDE Software Compilation, JuK, has been improved to simplify its use, with the ability to disable cross-fading and support for MP4 and ASF files (as long as the system has this support in Taglib). Michael Pyne and Jeff Mitchell implemented these improvements, taking advantage of some Kubuntu patches. [09:38] ah, I did those Kubuntu patches ^^ [09:38] * apachelogger is almost proud now [10:44] apachelogger for juk maintainer! [10:59] :D [10:59] now that might be a bit much === steveire_ is now known as steveire [11:28] Riddell: if it's too late to fix bug #564433 it should be a SRU [11:28] Launchpad bug 564433 in kdebase-workspace "[Lucid] New bug: Plasma crash at KDE logout" [Undecided,Confirmed] https://launchpad.net/bugs/564433 [11:31] debfx: is there a fix? [11:32] ah, yes, let's see if we can get that in [11:39] wow notify area goes away :) [11:39] ehlo peoples [11:46] ghostcube: hai [11:46] ghostcube: where is it going to? [11:47] http://design.canonical.com/2010/04/notification-area/ [11:47] into the orbit :D [11:47] hi apachelogger :) [11:48] i think this isnt the best way to go ... [11:49] but iam not someone have to decide anything heh [11:51] Riddell: Are you dealing with pitti's ping re kigo recommends? [11:51] way too long to read [11:51] did anyone else notice that lucid fills up the swap for no reason whatsoever? [11:52] apachelogger: xorg issue [11:52] -.- [11:52] horrible [11:52] ye, leak [11:53] braaargh I don't get this, why gold linker has so many failures here... [11:54] apachelogger: in short systray should go away [11:54] * Tm_T goes back to þe olde ld [11:54] no more task icons [11:54] one menu inside there are all icons categorized [11:54] and ubuntu want to push it out to the wild, no other distro so far [11:54] :) === jussi01 is now known as jussi [11:55] ScottK: yes [11:55] Great. I'll continue to ignore it then. [11:55] :D [11:55] * ScottK naps some more. [11:56] ahhh [11:57] so my system memleaks because they had to patch a bug that only affected silly clutter to begin with [11:57] sweet [12:00] ghostcube: honestly, from the things the post links to this is really just hiding things away in menus [12:01] and if so they only change the visual representation [12:01] yeah ... i dont get it really seems the only effect would be a not existing systray area [12:01] but in general it would be the same [12:03] "Many programs should not have an item in the panel at all. Where a notification area icon was being used mainly as a substitute for minimizing, the window should just minimize instead. We will be working on ways for long-running applications to be less obtrusive when their windows are minimized." [12:03] that however makes entirely sense [12:04] like say amarok, there is no point in having amarok in the tray other than preventing it from taking up loads of space if simply minimied [12:04] *minimized [12:06] agateau: I dont know what happened, but kopete doesnt load a context menu anymore and amarok doesnt show anything below the seperator (e.g. quit) [12:06] agateau: I also noticed that dbusmenu likes ot loose the stuff in amarok's context menu below the seperator a lot [12:07] like when the system is quite busy (say closing kontact and qt creator at the same time) it also doesnt load those items [12:07] right now I cant get them back though :( [12:08] hm [12:08] seems amarok doesnt even react to left click [12:08] ah [12:08] kopete neither [12:08] system is broken beyond repair [12:08] :D [12:08] well [12:08] yeah [12:09] I shouldnt click anywhere because it gets worse every time [12:10] apachelogger: many tray icons are for the wastebox :D [12:11] even the handling like right click left cclick middle click is a damn hazle [12:11] if they all would work with leftk klick like the autor said it would be an easier way [12:11] so every icon does what it wants [12:15] icons that do whatever they want are better implemented as independent applets/widgets, rather than jamming them into some arbitrary tray. [12:15] yep thats true [12:19] ghostcube: I really only now 2 kinds, those that have an associated window and those that do not [12:20] former would be quassel, latter knetworkmanager [12:20] former triggers window showing hiding on left click, latter mostly draws a menu (mostly the context menu) [12:21] middle click is undefined anyway [12:21] also I doubt a lot of users do middle clicking anyway [12:21] Many folks don't know that middle-clicking does anything. [12:22] Many laptop users *can't* middle-click (except by emulation) [12:32] JontheEchidna: is anything happening with bug 551456 ? [12:32] Launchpad bug 551456 in kcm-touchpad "systemsettings crashes when clicking "Keyboard & Mouse"" [High,Triaged] https://launchpad.net/bugs/551456 [12:40] apachelogger: yeah but if you have for example qjackctl as try icon for jackd it has an gui for left click and an menu on right click :) [12:41] ghostcube: and I did not mention that? [12:41] oh i should read all you type o.O [12:41] clearly a form of the first kind ... UI & context menu [12:42] and technically for the latter kind UI & context is also true, just that the latter kind does not have a window but limits it's UI to a menu type of thing [12:42] yeah [12:45] apachelogger: sorry was away [12:45] apachelogger: was your machine under heavy load when it happened? [12:49] agateau: yep [12:50] I am not sure if the lack of reaction was related to dbus menu though, maybe X just exploded [12:50] apachelogger: :) [12:50] the issue where amarok doesn't load all the entries is however reproducable every once in a while [12:51] in general dbusmenu exposes the usual oddness when under heavy load [12:51] apachelogger: lots of kde menus are created on the fly, [12:51] e.g. it also happens at times that the menu flickrs or that the menu is misplaced as seen in the bug earlier [12:51] apachelogger: to avoid locking the whole plasma ui, dbusmenu has quite short timeouts [12:51] apachelogger: which may be a problem under heavy load [12:52] apachelogger: but I am not sure what's worse: showing an incomplete menu, or waiting for a long time for the full menu to be ready [12:52] former IMHO [12:52] if the menu is under heavy load [12:53] it will be perceived at sluggish [12:53] so waiting for the menu to appear might seem more reasonable [12:53] an incomplete menu is more likely to be perceived as bug IMHO [12:54] even on heavy load [12:54] apachelogger: yes, you are right I think [12:57] Riddell: I've not been able to figure that one out :( [12:58] Riddell, debfx: Adding a m_brightnessOSD(0) to the Battery class initializer seems to fix the issue for me, though [12:58] (for the plasma-desktop crash) [13:00] debfx: ah, I see you already discovered that and attatched a patch to the patch :) === dendro-afk is now known as dendrobates [13:03] Hey guys just wanted to say congratulations this release of kubuntu seems to be a hell of a lot more polished, complete, functional etc etc etc [13:03] JontheEchidna: yeah sorry, should have pinged you [13:03] patching patches is really confusing ;) [13:03] debfx: actually I figured this out yesterday. I should have pinged you. ;) (Though I didn't catch the other variable not being initialized) [13:04] as long as it all works out in the end... :) [13:04] thanks davmor2 [13:04] davmor2: how's wubi installs doing? [13:05] the new brand for kickoff is not present in the RC, there is still the old one with kde logo [13:05] they all work one minor niggle with help me boot this cd but I'm not sure how many people will use that particular feature [13:07] Riddell: I am quite concerned about the kcm-touchpad crash, but I can't figure out a fix. [13:08] I really would like for us to be able to keep kcm-touchpad in the default install, but if necessary I suppose it may have to go if we can't find a fix [13:09] For maverick we may want to look in to using this KCM: http://kde-apps.org/content/show.php/synaptiks?content=114270 [13:09] It appears to be more functional. (And more importantly, maintained) [13:10] wasn't there new one in KDE 4.5? or was it that synaptiks just renamed, cannot remember === dendrobates is now known as dendro-afk [13:11] dunno, not heard anything abou that === dendro-afk is now known as dendrobates [13:14] seems to be in playground now, though: http://websvn.kde.org/trunk/playground/utils/synaptiks/ [13:28] KOffice 2.2rc1 tagged, yay === Tonio__ is now known as Tonio_ === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk [15:13] Riddell: no RC img of kubuntu-netbook for arm? [15:15] There wasn't? It passed RC testing: http://iso.qa.ubuntu.com/qatracker/build/kubuntu/all [15:17] persia: I think the arm images there are actually ubuntu ones [15:18] I don't know if kubuntu-netbook got tested [15:18] Oh. I keep misreading "netboot" as "netbook" :( [15:18] * persia goes to hunt someone with HW [15:19] ETIMEOFDAY :( [15:19] But no worries: if there are people testing, I'll make sure they sort out the ISO tracker for maverick. [15:20] freeflying: There's images on cdimage that you can use, containing the same software as the RC. There's a known bug for armel+omap, but the other subarches ought work. [15:21] freeflying: if you are able to test http://cdimage.ubuntu.com/kubuntu-netbook/ports/daily-live/20100420.1/ then I can ask for it to be published [15:24] dpm: do we have current output for the KDE vs Lucid language packs report? [15:24] persia: Riddell I see, thanks, going to download, hope cna finish testing by the end of tomorrow :) [15:25] Riddell, the last one I generated was the one I sent you, but I can update it. It might take a while, though, and would be best done for the final language packs being generated today. Let me come back to you later on. [15:26] hi [15:26] hi Adri2000 [15:27] I'm looking for the images to modify in order to customize the boot screen, kdm screen, session loading screen, and such. is it written in some place? or could some kind person help me? :p [15:28] Riddell, apachelogger, ScottK, I forgot to e-mail kubuntu-devel about the language pack deadline yesterday as I told Harald I'd do. Does a heads-up ping work for you, or do you still want me to send the e-mail although it's a day late? This is just about a reminder about the translations deadline having finished and the final language packs being generated (https://wiki.ubuntu.com/LanguagePackTranslationDeadline) [15:29] nixternal: what's happening with docs upload? [15:29] dpm: do you know why harald wanted the mailing list told? [15:29] Riddell: He's bitching and moaning and saying he can't upload it until Sunday. [15:30] * ScottK was about to ask the same question. [15:30] ScottK: hum, that doesn't leave much room for error [15:30] persia: so far, I don't have a omap's board :) [15:30] Adri2000: it's in various places, kubuntu-default-settings for plymouth screen, kdebase-workspace for KDM and ksplash [15:34] Riddell: I tried to point that out to him, but didn't get very far. [15:36] where is packagekit-qt upstream? I mean some source download possibility? [15:37] it's distributed with the rest of packagekit [15:38] in the libs/ directory of the packagekit source tarball [15:38] ah, thanks [15:38] wasn't clear to me [15:38] yw [15:38] apachelogger: you might be interested in this Plasma Desktop Scripting tutorial from aaron I'm scheduling http://www.doodle.com/im3gn9b9uu8acr65 [15:38] Riddell: ok, thanks! [15:42] Riddell, not sure, he just asked me to [15:46] apachelogger: bug 568965 [15:46] Launchpad bug 568965 in kubuntu-default-settings "Kickoff is not branded with new Kubuntu logo" [Undecided,New] https://launchpad.net/bugs/568965 [15:48] lex79: not a bug [15:48] no? why? [15:49] I thought we changed the icon in kickoff [15:50] that was the branding strip inside kickoff [15:50] not the kickoff icon itself [15:51] http://bazaar.launchpad.net/~kubuntu-members/kubuntu-default-settings/ubuntu/revision/386 [15:52] that was to support people who were optionally using the old-named icon [15:53] there is an option there but by default sticking to KDE artwork is the thing to do, KDE's intention is that distros use the kickoff branding bit [16:01] hi,I point out that,plasma-deskstop crash when ktorrent is active and turn off the computer [16:03] seawolf: we fixed a plasma crash on logout a few hours ago, bug 564433 [16:03] Launchpad bug 564433 in kdebase-workspace "[Lucid] New bug: Plasma crash at KDE logout" [High,Fix released] https://launchpad.net/bugs/564433 [16:04] ok,thank Riddell [16:07] wellllllllllllllllllllllllllllllllll [16:07] lex79: I tend to agree, just that for now we decided to stick with the KDE icon [16:08] the bug is a duplicate of another bug though [16:08] and that other bug is either fix released due to the icon being present or wontfix since we dont want to change the icon, at least not at this point [16:09] Riddell: so I can nail aaron with questions about the numerous short comings? ;) [16:09] apachelogger: exactly :) [16:09] perfect [16:10] * apachelogger is wondering what times he gets displayed [16:10] ok thanks apachelogger I hope we'll change the icon in Maverick :P [16:10] Riddell: I've got one of these new Dell laptops where the screen switching key is updated for Win7. Is there something we can do to make the key useful in Kubuntu? Like fire up krandrtray or something? [16:11] Riddell: mon 14 utc is the first slot, right? [16:12] ScottK: screen switching key? [16:12] apachelogger: yes [16:12] thanks [16:12] apachelogger: is that not clear? [16:12] Riddell: the UI doesnt say whats the time is against [16:12] which is a PITA in a cross-timezone scheduling [16:12] lex79 hi,I should tell you something,query ? [16:13] seawolf: kk [16:14] Riddell: in which channel? plasma? [16:14] Riddell: http://mjg59.livejournal.com/121851.html [16:15] lex79: haven't decided yet [16:16] ok [16:18] ScottK: that post seems to suggest there's no easy way to fix it [16:18] of course debfx is the king of the random laptop keys :) [16:18] Riddell: Yeah, he said KDE didn't currently make use of it. [16:19] So there's at least no conflict. [16:19] As it is, it just types a 'p'. [16:20] so just needs khotkeys set to run kcmshell4 display ? [16:22] I'm guessing, but I really don't know. [16:22] * ScottK tries [16:25] * ScottK concludes he doesn't get how to set it iup [16:25] iup/up [17:33] ScottK: amarok registers meta+p as toggle main window :/ [17:33] Ah. Well that's going to be a problem. [17:35] * Riddell updates kdevelop in beta PPA [17:37] ooh new koffice up, /me grabs that [17:41] who's up for a round of uno later on? [17:42] apachelogger: sorry, I'm too busy watching when snow is raining and when not [17:43] defenitely latter here :( [17:43] void Text::setText(std::string text) [17:43] { [17:43] text_ = text; [17:43] } [17:43] I just typedef std::string to text ^^ [17:44] s/just/should just [17:44] looks way better without that disturbing std::string === dendro-afk is now known as dendrobates [17:56] apachelogger: you could add "using std::string;" [17:56] I could [17:56] but that wouldn't help [17:56] it still would be a string amonst text [17:56] *amongst [17:56] or among? [17:56] something like that ;) [17:58] ah I see Text is a wrapper around std::string [18:02] debfx: not really [18:02] Text is part of namespace SVG, representing the SVG text shape :) [18:03] debfx: I was refering to the beauty of the pasted function, in only working with text of some sort [18:04] ScottK: we could easily add a khotkeys group that binds the display key and meta+p to show the display settings module [18:04] though I'm not sure what happens with a conflicting global shortcut [18:05] I think it would be really useful [18:05] True [18:05] debfx: whoever comes first gets it, unless the conflict is with something outside [18:05] in this case 2 actions will happen [18:06] s/something outside/anything that does not use KDE's shortcut magic [18:07] Well I think due to MS's Win 7 changes meta+p is going to be for video switching and it would be prudent for other apps to not use it. [18:11] Riddell: sunday [18:12] for docs upload === yofel_ is now known as yofel [18:54] * ScottK just discovered task bar thumbnails. Very nice. [18:55] debfx: display key? whats that? [18:56] maco: http://mjg59.livejournal.com/121851.html [18:58] maco: XF86Display, which is emitted when pressing some laptop function key [18:58] ahhhhh the key that doesnt work on my laptop that i was at one point trying to figure out how to fix [18:58] i backburnered it when i learned to use xrandr [18:59] well it probably works but just isn't bound to any action [18:59] it doesnt emit anything [18:59] the kernel has no idea that fn+f8 does anything at all [19:00] kde 4.4 helpfully pops up a box offering to let me configure screens *with a GUI* when i plug an external in. i always click "ignore" though because i trust my knowledge of xrandr more than i trust GUIs :P [19:00] p <--- That's what mine emits [19:00] mime shows amarok ^^ [19:01] *mine [19:01] * debfx has filed a bug about that === ramanK is now known as Guest26117 [19:01] amarok opens when you plug in a screen?? [19:02] or when you hit the fn thingy? [19:02] when I hit the fn key [19:02] Hi , when I click the kppp -> create new account , it crashes . does it a known bug? [19:03] is* [19:03] so all you need to do is get a bit of electrical tape the size of the top of your key, and draw a music note in white-out on it, and then TA-DA! it's a music key, not a display key [19:03] when I plug in a screen nothing happens. maybe because it's a VGA port? [19:03] VGA's all i've got [19:04] lucid detects when i plug in a screen. takes a second or two [19:04] Guest26117: define crashes please [19:05] apachelogger: it shows a blank dialog box and does nothing [19:05] * apachelogger is wondering why his kppp is chowned to root:dip actually === dendrobates is now known as dendro-afk [19:05] Guest26117: are you on karmic? [19:06] apachelogger: I'm on lucid + latest upgrades [19:06] that is strange because it does not start here :P [19:06] oh [19:06] my [19:06] apachelogger: it starts only by root [19:07] * apachelogger ignores the family friendlyness and shouts FUCK [19:08] * Tm_T looks angry at son [19:08] ScottK, Riddell: kppp is the broken out of the box [19:08] can we pushy pushy update for that before final? [19:09] Guest26117: the thing is that you mustn't start it as root [19:09] Guest26117: sudo chown root:dialout /usr/bin/kppp [19:10] that should make it startable without root for the time being [19:10] apachelogger: How invasive is the fix? [19:10] * maco raises an eyebrow at apachelogger [19:10] apachelogger: thanks , i'll do it [19:10] ScottK: chown root:dialout /usr/bin/kppp [19:11] apparently I forgot to change debian's dip to dialout in the kdenetwork rules [19:11] dip is no default group on ubuntu though [19:11] so kppp will not even start by default [19:12] Nice. [19:12] I asked the RM if we can get it in. [19:12] apachelogger: now it says : Configuration file "/home/raman/.kde/share/config/kppprc" not writable. [19:12] Please contact your system administrator. [19:12] !!! [19:12] !!! [19:13] dont start gui apps with sudo [19:13] Guest26117: chown -Rv $USER:$USER ~/.kde [19:13] apachelogger: I started it by kick-off menu [19:13] apachelogger: You have release manager approval to fix it. [19:13] Guest26117: kickoff menu does not start it with sudo, does it? [19:13] apachelogger: Please upload as soon as you have it working. [19:14] ScottK: thanks [19:14] Guest26117: you started it with sudo one time before, right? [19:14] apachelogger: no , it doesn't [19:14] Guest26117: that changed the permissions on the files in your ~/.kde [19:14] maco: yes , I did [19:14] point in case :P [19:14] apachelogger: you got that a bit backwards... [19:16] maco: yeah, or maybe it is just like I always say weird when I am talking about wired ;) [19:16] apachelogger: you are weird, that's for sure [19:16] ...wired? [19:20] Tm_T: hungry [19:20] I have another problem on lucid : [19:21] If I manually suspend to RAM it works perfectly , but if I do it by closing the lid , after opening it , it takes several minutes to show the password dialog box [19:22] Guest26117: does kppp work now? [19:22] apachelogger: thanks , it works perfectly [19:22] Guest26117: There's a known X server memory leak. A fix is coming before release. [19:22] That would explain that [19:22] ScottK: interesting [19:23] [kdenetwork] Harald Sitter * apachelogger@ubuntu.com-20100423182354-2w2esfw4z9ihd3jo * debian/ (changelog rules) Actually change the debian/rules file as promised in ubuntu2 (LP: #278408) [19:23] Launchpad bug 278408 in kdenetwork "KPPP blank/unresponsive "Create New Account" dialog" [High,Fix released] https://launchpad.net/bugs/278408 [19:24] ScottK: uploaded [19:24] Guest26117: cool, thanks [19:24] ScottK: ooooh really??? [19:24] ScottK: could this X server memory leak explain why i often have all 4GB of memory in use after a couple days running? [19:25] ScottK: current daily powerpc image going to testrun [19:26] And my last problem in lucid is this : https://bugs.kde.org/show_bug.cgi?id=234749 [19:26] KDE bug 234749 in widget-battery "battery monitor shows the AC adapter is plugged in at startup , but it's not" [Normal,Unconfirmed] [19:27] maco: Could. [19:28] Tm_T: Great. [19:28] * ScottK doesn't have a good answer for that last one [19:28] :( [19:31] kttsd don't work ,I feel spd-say hello, but I hear no sound [19:35] does anybody know if the nvidia prop. drivers could do grayscale well for plymouth? [19:38] My OOWriter sometimes does not show the icons (such as zoom , page view , ... ) at status bar , does yours do it? [19:41] apachelogger: Accepted. Thanks. [19:41] Guest26117: The kppp fix has been accepted for Lucid and is in the queue for building. [19:41] Guest26117: known bug [19:42] ScottK: great , thanks [19:42] maco: thanks for informing [19:42] Just keep this in mind the next time you hear someone whining about non-responsive developers. [19:42] ScottK: ;) [19:42] hahah [19:43] Where is the control knob for panel transparency? [19:43] i thought that was theme set? [19:43] BTW, we lost the "are you sure you want to remove the panel" warning patch somewhere along the way. [19:44] Maybe that's why I can't figure out how to change it. [19:45] i think you're thinking of gnome [19:45] it has a way to set the panel's transparency separate from everything else [19:45] * maco has read the source code of that [19:45] The here atmosphere is awsome , I'm a c++ programmer and I know a little about qt , is there any way that I can help ? Is any help needed? [19:46] Guest26117: fix last minute bugs before release? [19:46] What can I do? [19:47] maco: At this point most things will go into post-release updates. The reason the kppp fix got in was broken dial up networking could interefere with one's ability to get updates. [19:47] gotcha [19:47] bug 551456 could use some attention from a C++ whiz [19:47] Launchpad bug 551456 in kcm-touchpad "systemsettings crashes when clicking "Keyboard & Mouse"" [High,Triaged] https://launchpad.net/bugs/551456 [19:48] JontheEchidna: Good point. [19:48] ScottK: should be done upstream anyway ... re warning patch [19:48] didnt crash for me... [19:48] apachelogger: Agreed. [19:48] I am sure usptream finds a way to turn it into an approach that breaks within 2 releases time :P [19:48] apachelogger: I wasn't complaining, just informing [19:49] * apachelogger spent like 5 hours today writing mails about design stuff for a c++ exercise -.- [19:50] from what I can see, the crash is happening in Touchpad::get_parameter() in touchpad.cpp, and the crash happens when the function is not able to grab the parameter due to permissions [19:51] oh, hey. that's a void class. Why is it returning anything at all? [19:51] hrmhrm [19:52] because it's a pointer to a void function? [19:52] ew, this is some trippy code [19:52] Guest26117: ^^^ [19:53] sort've requires a laptop to debug, though [19:55] I tried the touchpad settings , it works without any issue for me [19:56] when a user is not in the admin group, it crashes on startup :( [19:56] so if you created a second user and tried to open up the module -> fail === ramanK is now known as Guest45281 [20:03] I created a new user and clicked the touchpad setting but it works correctly for me [20:04] but another problem occured ! when I clicked the switch user , and then the new session , it showed a black screen in died!!! [20:04] hmm, interesting [20:05] on both counts ;) [20:05] :) [20:06] I'll try it again [20:07] wow, syaptiks is pretty spiffy. It even has documentation [20:07] lol [20:07] docs ftw! [20:08] I may definitely have to advocate synaptiks over kcm-touchpad for next cycle [20:08] JontheEchidna and Guest45281: I have another clue for you: The touchpad crash isn't just if the user doesn't have admin rights, it needs to be no user with admin rights is logged in. [20:08] * ScottK got the black screen thing too [20:08] aah [20:09] if we are voting on stuff, i say shaman over kpackagekit, i hate kpackage kit [20:10] shaman doesn't have a usable dpkg backend [20:10] it useless for anything other than the very basics [20:10] nor is its UI anywhere close to being usable [20:10] JontheEchidna: that doesnt matter! [20:10] JontheEchidna: I think kpackagekit and shaman tie on usablility [20:10] chakra uses it! [20:10] chakra is cool and awesome [20:10] :P [20:10] thy must advocate the shaman I say [20:11] how shaman is supposed to work, btw? [20:11] I fail with it [20:11] BTW, I think I am finally motivated to learn enough GUI programming to get sensible update/restart notifications back. [20:11] That may be the only thing I do the next cycle. [20:11] ^^ [20:11] ScottK: well, now that the design team removes the notification area... :P [20:11] :P [20:11] shaman is too buggy , it killed my chakra just by a normal upgrade [20:11] apachelogger: Fortunately we are Ubuntu, not Kubuntu [20:11] also I read they have intrusive popups of the update window [20:12] Strike that, reverse it. [20:12] like have the window itself popup [20:12] apachelogger: Yes. They believe the solution is to hide updates from the user. [20:12] imagine how annoying that would be :D [20:12] ScottK: that is not what the blog post about systray-be-gone said [20:12] It did. [20:12] the 25% I actually read that is ^^ [20:13] Well it didn't [20:13] oh my, xorg is recognizing my laptop's touchpad as a wheel mouse now :/ [20:13] On their mail list they said they want to announce updates less often [20:13] no touchpad config for me :( [20:13] * apachelogger installs stuff for linuxtage tomorrow [20:14] ScottK: that makes sense depending on how often they want [20:15] there is no point in telling the user 3 times a day that a random update appeared [20:15] for security maybe [20:15] but sure not for regular updates [20:15] apachelogger: I think that if your update notification is so disruptive you need to consider not letting people know about updates you are solving the wrong problem. [20:15] * JontheEchidna notes that packagekit doesn't listen to the apt update-frequency setting at all [20:15] well, it is contributing to business of the system [20:16] * apachelogger is all for zen desktops TBH [20:16] then again I am also using flux at times [20:17] Nightrose: any clues on that weird naming scheme that implies linuxtag is always longer than one day and linuxtage are one day? [20:17] lol [20:17] a study colleage turned that into a running gag the other day [20:17] no idea === dendro-afk is now known as dendrobates [20:18] Nightrose: "gehst du zu den linuxtag?" ^^ [20:18] Nightrose: when is that anyway? [20:19] kubotu: google linuxtag berlin [20:19] Results for linuxtag berlin: 1. LinuxTag 2010: http://www.linuxtag.org/ | 2. LinuxTag 2010: http://www.linuxtag.org/2010/en.html | 3. Linux Tag - Berlin | Spread Firefox: http://www.spreadfirefox.com/node/3775 [20:19] apachelogger: nope - got akademy and community leadership summit coming up in july already [20:19] so no linuxtag for me [20:19] * apachelogger seems to remember that he cant go there for some reason [20:20] Nightrose: you are too busy :( [20:20] i am indeed [20:20] too much work and no time for apachelogger makes Harald sad :( [20:20] but so looking forward to a trip to finland and the us [20:20] awwww [20:20] *hug* [20:20] *rehug* [20:20] Finland... should make a trip there, yes [20:21] ah [20:21] oh wait [20:21] the dictator is keynoting at linuxtag [20:21] Tm_T: time for your meds [20:22] oh cool bill hilf is also keynoting [20:22] :/ [20:22] I probably should consult my calendar again [20:23] Tm_T: How's powerpc? [20:24] ScottK: still burning [20:25] Tm_T: They have pills for that now. [20:25] lovely <3 [20:29] Time to try the new xorg. [20:31] New laptop has ssd, so that didn't take long. [20:32] my iPaq uses sdram as mass storage memory of some sort [20:38] ScottK: rebootinh to live [20:39] Excellent [20:45] it looks like kcm-touchpad doesn't really do error checking [20:45] Touchpad::get_parameter() does some checks and returns NULL on error but the result is dereferenced anyway [20:47] I really don't understand how a void function could return anything at all [20:47] and still compile [20:47] i thought void functions could return anything they wanted without it being an error [20:48] while if you dont return on a non-void *then* its an error [20:48] void ends up acting like wildcard [20:48] youre supposed to assume it returns nothign when you call it [20:48] (i think one of my professors mentioned this recently and surprised me) [20:50] maco: void also doesn't care if you don't return a thing [20:50] it's, err, void [20:50] right, wildcard [20:50] * can match 0-infinity characters in a regex [20:50] doesn't sound too safe to return from a void function, at any rate [20:50] agreed [20:51] the calling function shouldnt assume anything at all about what's returned because it *could* be anything [20:52] lol: http://learnfrommyfail.com/2010/02/05/fail-story-lfmf-isa/ [20:52] Nice [20:54] uh yeah agree with "omg who throws away chocolate?" and "chunky stuff goes in trash if you lack disposal" [20:54] JontheEchidna: get_parameter returns void* [20:54] wait void pointer not just void? [20:55] yes [20:55] so I was right when I said it returned a pointer to a function? [20:55] returning null in that situation would not be good [20:55] insta-segfault [20:55] are you talking about get_parameter()? [20:56] yeah [20:56] it doesn't return a function pointer [20:58] then does it returns a pointer to a wildcard value of whatever it wants? (e.g. the return value of another function?) [20:58] yes [20:59] O_o [20:59] is it at least always the *same* function's return value? [20:59] if not: OH MY [20:59] of course not ^^ [21:00] it looks to me like it's either returning dp_get_parameter or NULL [21:00] NULL In the case where display && device is not true [21:01] dp_get_parameter(display, device, name); otherwise [21:01] really good coding style: (int)*(char*)Touchpad::get_parameter(...) [21:03] O_o [21:03] kill it with fire! [21:03] (it's the only way to be sure) [21:04] Who's idea was this kcm? [21:04] * JontheEchidna flees [21:05] Synaptiks appears to have a much cleaner codebase. (And in fact, looks to be better all around in terms of functionality, UI, docs and such) [21:05] the kcm-touchpad author also seems to have disappeared... [21:06] I really must flee now though, time to go home. ;) === dendrobates is now known as dendro-afk [21:08] how can I reproduce the crash? starting it as non-admin user doesn't work for me [21:09] debfx: Did you switch user and leave an admin user logged in? [21:09] You need to log the admin user out [21:10] no, I logged in as non-admin directly after booting [21:10] hmm, not leaving until 5 today. ok [21:10] It appears OOo does violence on read only sections of word documents in Lucid. Does anyone have a document with such sections that could go in a bug report? The only ones i have aren't distributable. [21:19] Did we ever get any good data on where we are on the translations suck-o-meter for this release? [21:20] ScottK: do they have to be made by MS Word or can OOo make a read only bit save as .doc and then fubar it when re-opened? [21:21] maco: Dunno. The only ones I have were originally made in word [21:22] Most languages should be fairly ok compared to upstream, as far as we know [21:22] Lithuanian is screwed, though: bug 565294 [21:22] Launchpad bug 565294 in rosetta "Plural translations are broken for Lithuanian language" [High,In progress] https://launchpad.net/bugs/565294 [21:24] (this is something that wouldn't be a problem if we didn't use rosetta, for the record) [21:24] But without multilingual people telling us the quality of changed translations in LP versus upstream, there's no way to know how good they are [21:25] instead we'll either only get the reports after final release, or they'll just chalk it up to "kubuntu sucks" and never report :/ [21:26] * maco wishes she was more fluenter at being multilingual [21:27] heh [21:33] on the plus side, we have more langpacks on the cds than we have had in a long while [21:36] does kcm-touchpad print anything on stderr before crashing? [21:37] I've never been able to reproduce, as xorg doesn't detect my touchpad as a touchpad, and doesn't load the synaptics driver for it :( [21:39] I can reproduce it if you need information. [21:46] ScottK: could you give me the console output and a stacktrace (with kcm-touchpad-dbgsym installed) [21:46] OK [21:47] What's the sources.list line for the dbgsym repo? [21:48] deb http://ddebs.ubuntu.com/ lucid main restricted universe multiverse [21:48] Thanks [21:49] * ScottK bets the package install goes better after he adds that [21:57] heh [21:58] What's the command to run kcm-touchpad from konsole? [21:58] kcmshell4 --list tells the items [21:59] then kcmshell4 [22:00] Just starting systemsettings seemed to do it. [22:00] And crash .... [22:01] ScottK: dpm said he'd run the comparison tool after the language packs get built today [22:01] OK. Sounds good [22:02] debfx: Traceback: http://paste.ubuntu.com/421278/ [22:03] debfx: Konsole output: http://paste.ubuntu.com/421279/ [22:04] i got my brother to upgrade to lucid a few days ago :) [22:04] maybe i'll corrupt him into a beta tester next time round instead of an RC tester.... [22:05] Riddell: Was this tool run earlier in the cycle too? [22:05] ScottK: yes although I only saw output for one language [22:10] OK. [22:11] It would have been nice to have some trend information. [22:11] ScottK: thanks [22:11] You're welcome. [22:20] ScottK: does "synclient -l | grep VertEdgeScroll" output a format mismatch? [22:20] debfx: I can't switch users right at the moment. I'll check shortly. [22:33] debfx: Yes [22:33] debfx: http://paste.ubuntu.com/421297/ [22:33] debfx: Do you need anything else before I switch users? [22:34] ScottK: don't think so === dendro-afk is now known as dendrobates [22:34] OK. Great. === dendrobates is now known as dendro-afk [22:53] kubotu: order coffee for Nightrose [22:53] * kubotu slides coffee with milk down the bar to Nightrose. [22:53] yay [22:53] very much needed [22:53] :) [22:53] i'm drowning in paper proposals [22:53] and need to put them all in a spreadsheet [22:53] ohmy [22:54] Nightrose: the KDE community is in need of a web 2.0 application to explore the full potential of proposal evaluation [22:55] very much so [22:55] can't be that difficult really, have a stict UI that commits a proposal to a database, then have a simple query page for the database [22:56] that also makes processing a lot easier, if the submission UI is strict enough to enforce some sort of structure onto the submssion [22:56] indeed [22:56] email is quite the mess [22:57] especially if a lot of people don't know what an abstract should look like [22:59] * apachelogger notes that one could also do Qt apps that submit xml files to ftp, then have another app fetch that ^^ [22:59] we tried to use whozzname for akademy when we had it in scotland [22:59] since no one wants to use the php [22:59] to submit papers [22:59] but the server was broken on the day and nobody has trusted it since [23:00] * apachelogger notes that this is why xml + ftp might be better ^^ [23:01] Nightrose: going to froscon? [23:01] apachelogger: if workwork sends me yes [23:01] i've suggested it would be a good idea to do so today ;-) [23:02] hehe :D [23:02] * apachelogger is thinking about proposing a talk [23:02] as usual I dont know about what though ^^ [23:02] also I have no clue when the cfp ends [23:03] apachelogger: one hour to go (in my timezone) submit quick! [23:04] Riddell: not akademy, Nightrose is already under so much load :) [23:04] + I think we already established that I cannot attend akademy for some reason [23:04] apachelogger: one more can't break me... [23:04] ;-) [23:05] apachelogger: ftp is never better. [23:05] 3-10th July is about exam time [23:05] ScottK: ftp never ever breaks the servers [23:06] well, I never saw that happen though [23:06] ftp is so 1992. [23:06] it regularly happens that some db (and by that I mean the devil of relational databases) takes down a serve [23:06] ftps, sftp, or scp at least please. [23:06] point being? [23:07] Point being don't use ftp. [23:07] sftp is ftp tunneled through a ssh channel, isnt it? [23:07] Mostly. [23:07] it's not like you would submist most stuff via ssl either ;) [23:07] They key point being the through the tunnel. [23:07] especially not if the devial of relational databases is backing the web UI [23:07] Friends don't let their friends use FTP. [23:07] ^^ [23:08] well, so be it some other old protocol [23:08] we could also 777 on some http server ;) [23:08] or send by mail after all [23:09] though kmail is not as dynamic in viewing the data as a custom UI would be [23:09] * apachelogger notes that krake disappeared and didnt tell apachelogger when to be where ^^ [23:29] lol [23:29] nixternal: http://blog.nixternal.com/2010.03.23/desktop-help-summit-2010-pictures-available/ havent seen that one before [23:29] <3 [23:42] nixternal: i showed that pic to one of my old profs and he goes "yes, kde IS the best. do you use kde yet?" and i pointed to kickoff and said "what does that look like?" and he goes "you DO! when did you switch?" [23:55] and then he complained that there's no plasmoid in the panel to use as a konsole like in kde3. i found one called plasmacon but oh goodness does it fail horribly inside the panel